R programming: packages

M1 MIDS/MFA/LOGOS

Université Paris Cité

Année 2024

Course Homepage

Moodle

This lab is just an incentive to dig into R Packages (2e) by Wickham and Bryan

Introduction

Sharing/Storing/Organizing

Packages come with conventions

Packaging can be helpful to perform data analysis.

Setup

Automating package development is facilitated by a collection of packages.

stopifnot(
  require("devtools"),
  require("usethis"),
  require("testthat"),
  require("styler"),
  require("roxygen2")
)

devtools is the cornerstone of the collections

• with rstudio (hands in hands)
• with vs code
• with positron

Package organization

Make sure you understand the different states of a package : source, bundled, binary, installed, in-memory.

• At first, we deal with a source package. Where is it hosted?
• How is the source organized?
• What are metadata for?
• What is the purpose of file .Rbuildignore?
• Why do we bundle packages?

Structure and states of a package

To initialize a package development directory.

packpath <- "hmw2.ma7by020"

if (!fs::dir_exists(packpath)){

  fs::dir_create(packpath)
}

usethis::create_package(packpath)

Within directory hmw2.ma7by020, you should have the next organization

hmw2.ma7by020/
├── DESCRIPTION
├── .gitignore
├── hmw2.ma7by020.Rproj
├── NAMESPACE
├── R
└── .Rbuildignore

2 directories, 5 files

What is the difference between loading and attaching a package?

You will have to

• populate the R subdirectory
• update DESCRIPTION

Reusing scripts

Styling

Testing

The workhorse of the development process is

devtools::load_all()

1. Code and/or fix bugs
2. Load the code
3. Test the code

Checking Package state

devtools::check()

NAMESPACE and dependencies

Versioning (git)

References

Writing R extensions